Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Jie Chen Project Name: ScalaGAUSS Division: MCS Project title: Scalable Gaussian Process Data Analysis Associated funding: Spatio-Temporal Data Analysis at Scale Using Models Based on Gaussian Processes Scalable Gaussian Process Analysis of Spatio-Temporal Data (PI: Mihai Anitescu) Other Systems: MCS workstations Science: This project uses a maximum likelihood approach to perform Gaussian process analysis on both synthetic and real data. The observations considered for analysis are simulation and data from nuclear engineering and climate science. The resulting analysis is used to describe the spatial correlation of the sampled data, to predict and to understand uncertainty in very high fidelity phenomena. Project description: ScalaGAUSS is a statistical software for analyzing Gaussian process data arising from various applications (climate, material science, nuclear). The goal of ScalaGAUSS is to perform scalable computations and handle large data sets in the order of 10^9 and beyond. The development of ScalaGAUSS is based on both the innovation of numerical algorithms and the implementation of parallel programs. We have demonstrated in our FY2014 report the ability to handle data in the scale of 10^9 with good scaling to 1024 processes. Such capability is, however, limited by the special data configurations (regular grid) or the particular covariance kernels (Matern kernel). As a continuing effort, we have been developing a novel numerical methodology, coined “recursively low-rank compressed matrices”, for handling the most general case (scattering data and arbitrary kernel). We are requesting allocations for the following work in FY2015: 1. Wrap up the ongoing experimentations and furnish technical papers to be submitted to conferences and journals. 2. Carry out the code development, testing, and production runs for the new methodology, recursively low-rank compressed matrices. 3. An LDRD project, entitled “O(n) Dense Matrix Algorithms and Software for Simulation and Data Analysis”, is under the second stage review. The technical content of the LDRD project is aligned with this new methodology. If the LDRD project is funded, the Blues cluster is an important computational resource for the development of the methodology. The above work requires similar number core hours as in the past year. Industry partnership: Project URL: http://press3.mcs.anl.gov/scala-gauss/ Current FY Hours Used: undetermined amount New FY Requested allocation: 480000 Q1: 120000 Q2: 120000 Q3: 120000 Q4: 120000 Justification: Storage requirements: Thank You, The LCRC Accounts System